Fix/neural docs #473
Fix/neural docs #473
Conversation
docs/neural/index.rst
Outdated
| .. module:: ott.neural | ||
| .. currentmodule:: ott.neural | ||
|
|
||
| Under reconstruction. .. TODO(marcocuturi): add some nice text here please. |
There was a problem hiding this comment.
@marcocuturi can you please add something here?
There was a problem hiding this comment.
wasnt sure you wanted it now :)
In contrast to most methods presented in :mod:~ott.solvers, which output vectors or matrices, the goal of the :mod:~ott.neural module is to parameterize optimal transport maps and couplings as neural networks. These neural networks can generalize to new samples, in the sense that they can be conveniently evaluated outside training samples. This module implements layers, models and solvers to estimate such neural networks.
docs/neural/solvers.rst
Outdated
| .. module:: ott.neural.solvers | ||
| .. currentmodule:: ott.neural.solvers | ||
|
|
||
| Under reconstruction. .. TODO(marcocuturi): add some nice text here please. |
There was a problem hiding this comment.
@marcocuturi also here, thank you!
There was a problem hiding this comment.
This module implements various solvers to estimate optimal transport between two probability measures, through samples, parameterized as neural networks. These neural networks are described in :mod:~ott.neural.models, borrowing lower-level components from :mod:~ott.neural.layers, building from the
flax <https://flax.readthedocs.io/en/latest/examples.html>__ library.
Codecov Report
Additional details and impacted files@@ Coverage Diff @@
## main #473 +/- ##
==========================================
+ Coverage 90.41% 90.47% +0.05%
==========================================
Files 59 60 +1
Lines 6440 6541 +101
Branches 912 928 +16
==========================================
+ Hits 5823 5918 +95
- Misses 472 478 +6
Partials 145 145
|
* Remove nn from docs, fix some small issues * Start adding neural part * Update neural docs * Fix spellchecker * First round of comments * Fix TODOs
Fixes the
ott.neuraldocs. Also fixes some minor typos and not correctly forwarding argument inott.neural.losses.monge_gap.@marcocuturi I left 3 small todos to update some of the description in the docs.